Modern MBA

Case study — Services · 7 min read · 5 questions

Why almost nobody makes money in an MLM

The thesis

In a normal business you make money selling to customers. In an MLM you make money recruiting people who become your customers — and once the seller and the buyer are the same person, revenue can be manufactured rather than earned.

That works, for a while, because the model runs on information arbitrage. It needs people who cannot easily compare prices, cannot find the product elsewhere, and have no better option for flexible income. Every one of those conditions has been destroyed by the smartphone: price transparency, next-day delivery, and rideshare apps that pay more reliably than a downline ever did.

The compensation flywheel is the trap. Once it is switched on it cannot be switched off, because the payouts are what hold the sales force together. The company must keep feeding new products, new incentives and new markets into it — which means it can never redirect resources to the thing that would actually save it. Every MLM ends up captured by its own sellers.

How do you think about this? 5 strategy questions this case raises and does not answer.
Read the comments, or add yours →

The statistics

$15.5B → $0Avon market value, 2010s to today
90% vs 1%Herbalife sellers at the bottom vs the top
1 in 10,000Herbalife distributors who reach the top tier

By the numbers — swipe or use arrows

01The graveyardMarket value of the major MLMs by era, in billions. Not a downturn — an extinction, across every company in the category at once.
The graveyard — Why almost nobody makes money in an MLM$0B$5B$10B$15B$20B$15.5B$0BAvon$8.2B$1.2BHerbalife$5.7B$0.2BNu Skin$4.8B$0BTupperware$3.1B$0.1BMedifast$1.3B$0.3BUSANA2010–20142026Modern MBA
View data
2010–20142026
Avon$15.5B$0B
Herbalife$8.2B$1.2B
Nu Skin$5.7B$0.2B
Tupperware$4.8B$0B
Medifast$3.1B$0.1B
USANA$1.3B$0.3B
02What $1,000 becameAverage value of $1,000 invested in MLM stocks by era. The 2015–2019 cohort looked like a winning trade right up until it wasn't.
What $1,000 became — Why almost nobody makes money in an MLM$0$2,000$4,000$6,000$2,378$700Herbalife$2,086$252Nu Skin$5,183$1,424USANA$1,000$0Avon$1,493$0Tupperware$5,746$723Medifast2015–20192026Modern MBA
View data
2015–20192026
Herbalife$2,378$700
Nu Skin$2,086$252
USANA$5,183$1,424
Avon$1,000$0
Tupperware$1,493$0
Medifast$5,746$723
03Where the sellers landHerbalife's own distribution of sellers by tier. Nine in ten never leave the bottom, and the shape does not change from year to year.
Where the sellers land — Why almost nobody makes money in an MLM0%25%50%75%100%90%9.2%201190%8.8%201290%8.6%2014BASEMIDDLEModern MBA
View data
BaseMiddle
201190%9.2%
201290%8.8%
201490%8.6%
04And what they grossAverage annual gross for Herbalife's top tier against everyone else. The spread is the recruitment pitch — and it is funded from below.
And what they gross — Why almost nobody makes money in an MLM$0$100,000$200,000$300,000$400,000$259,511$3,0002011$284,680$3,0002012$311,466$3,0002014TOP TIERBASE TIERModern MBA
View data
Top tierBase tier
2011$259,511$3,000
2012$284,680$3,000
2014$311,466$3,000
05The same shape everywhereTupperware's seller distribution across fifteen years. Different company, different decade, identical proportions — because the structure produces them.
The same shape everywhere — Why almost nobody makes money in an MLM0%25%50%75%100%95%0.8%200596%0.9%200994%1.2%2018CONSULTANTSDIRECTORS AND ABOVEModern MBA
View data
ConsultantsDirectors and above
200595%0.8%
200996%0.9%
201894%1.2%
06The flywheel eats the companyTupperware's commissions and incentives as a share of sales. Holding the sales force together consumed more than half of every dollar by 2019.
The flywheel eats the company — Why almost nobody makes money in an MLM0%20%40%60%49.8%200749.8%200949.6%201150.2%201350.6%201551.2%201755.6%2019Modern MBA
View data
Value
200749.8%
200949.8%
201149.6%
201350.2%
201550.6%
201751.2%
201955.6%
07Tupperware's long erosionRevenue and operating income through the decades. The peak was 2013, and the profit had already gone before the revenue did.
Tupperware's long erosion — Why almost nobody makes money in an MLM$0M$1,000M$2,000M$3,000M$1,045M$111M1999$1,114M$87M2003$1,983M$170M2007$2,584M$331M2011$2,671M$386M2013$2,256M$289M2017$1,797M$67M2019REVENUEOPERATING INCOMEModern MBA
View data
RevenueOperating income
1999$1,045M$111M
2003$1,114M$87M
2007$1,983M$170M
2011$2,584M$331M
2013$2,671M$386M
2017$2,256M$289M
2019$1,797M$67M
08Always a new marketTupperware revenue by region. As the West caught on, growth had to be found somewhere else — the only move a pyramid has left.
Always a new market — Why almost nobody makes money in an MLM$0M$250M$500M$750M$1,000M$546M$223M2003$688M$292M2007$849M$714M2011$783M$840M2013$605M$780M2015$475M$591M2019EUROPEASIA PACIFICModern MBA
View data
EuropeAsia Pacific
2003$546M$223M
2007$688M$292M
2011$849M$714M
2013$783M$840M
2015$605M$780M
2019$475M$591M
09The settlementHerbalife's US segment before and after the FTC tied rewards to verified retail sales. Revenue held, then decayed once the pandemic bump passed.
The settlement — Why almost nobody makes money in an MLM$0M$500M$1,000M$1,500M$841M$284M2012$926M$312M2014$956M$301M2016settlement$1,320M$433M2020$1,226M$375M2022$1,033M$308M2025REVENUEOPERATING INCOMEModern MBA
View data
RevenueOperating income
2012$841M$284M
2014$926M$312M
2016 settlement$956M$301M
2020$1,320M$433M
2022$1,226M$375M
2025$1,033M$308M
10The activist tradeHerbalife's share price through the Ackman short, the FTC investigation and the settlement. The company survived both the attack and the argument.
The activist trade — Why almost nobody makes money in an MLM$0$25$50$75$100$34Dec 2012Ackman$40Q1 2013Icahn$50Q2 2013$79Q4 2013$49Q1 2014probe$38Q4 2014$54Q4 2015$65Jul 2016settlement$92Feb 2018Ackman exitsModern MBA
View data
Value
Dec 2012 Ackman$34
Q1 2013 Icahn$40
Q2 2013$50
Q4 2013$79
Q1 2014 probe$49
Q4 2014$38
Q4 2015$54
Jul 2016 settlement$65
Feb 2018 Ackman exits$92
01 / 10

Figures from company 10-K filings and income disclosure statements, published compensation plans, FTC and SEC enforcement records, and Pershing Square Holdings research on Herbalife

Key takeaways

01

The category did not decline, it was wiped out. Tupperware went bankrupt in 2024, Avon was dissolved the same year, NewAge collapsed two years earlier, Medifast has fallen 96% from nearly $300 to $11 a share, and Herbalife is down over 80%.

02

Avon was worth $15.5 billion in the early 2010s and is worth nothing today. Tupperware went from $4.8 billion to $0, and USANA from $1.3 billion to $0.3 billion — an entire industry's market value erased inside fifteen years.

03

The earnings distribution is the business model, not a side effect. At Herbalife, 90% of sellers sit in the base tier, 8.6% in the middle, and 1% at the top — and the company's own filings put the odds of reaching the top tier at roughly 1 in 10,000.

04

At Tupperware the split was almost identical across fifteen years: 95% consultants in 2005 and 94% in 2018, with the top tier never exceeding 1.2% — stable proportions are what a pyramid produces by design.

05

The gap between tiers is the recruitment pitch. Top Herbalife sellers grossed over $311,000 while the base tier grossed a few thousand — a spread that only exists because the money flows up from recruits rather than out from customers.

06

The cost of holding the pyramid together rises until it eats the company. Tupperware's commissions and incentives climbed from 49.8% of sales in 2007 to 55.6% by 2019 — more than half of every dollar spent keeping sellers from leaving.

07

Regulators priced the practice rather than stopping it: the FTC fined Herbalife $200 million and forced it to tie rewards to verified retail sales, Avon paid $135 million over bribery charges, and Nu Skin was fined in China — and the settlements changed the numbers more than the model.

08

Geographic expansion is the tell. When a market catches on, the only move is a new one — which is why Tupperware stayed in Venezuela through hyperinflation, and why seven of every ten dollars eventually came from outside the West.

09

Tupperware's original advantage was social, not commercial. The host controlled the venue, hand-picked the guests and pocketed 25–35% of every container sold — a sales environment no retailer could replicate and no competitor could enter.

10

Herbalife's Mexican reinvention is the most impressive thing any MLM has done. Low-income buyers could not afford $120 tubs, so sellers split freshly mixed shakes into single servings for $2–3 in their living rooms — and by the early 2010s there were over 20,000 nutrition clubs in Mexico alone.

11

That pivot rebuilt the whole company around a different customer. Tupperware rode the 1950s suburban housewife; Herbalife rode the 21st-century Latino consumer, launching hundreds of products a year by market — high-protein iced coffee for Americans, rice-pudding flavours for Mexicans, kulfi shakes for Indians.

12

It worked well enough to make it the category's champion. By 2015 Herbalife was the biggest MLM on the American stock market, and it still earns billions a year — while the stock has fallen more than 80% over the past decade.

13

Inflation is now eating the innovation that saved it. A daily $9 shake-and-tea combo has gone from an affordable routine to a luxury for exactly the working-class customers the nutrition clubs were built for.

14

The finding was about the promise, not the product. The FTC's case was that Herbalife marketed selling as a path to financial freedom while those in the middle and bottom earned on average under $5 a month — and the settlement required at least 80% of US sales to go to genuine retail customers.

15

Medifast shows the same collapse in the most exposed category. Selling weight-loss coaching and meal replacements to desperate dieters, its valuation evaporated 96%, from nearly $300 to $11 a share.

16

The strongest argument against the model is that nobody else copies it. If dangling money and a rags-to-riches story genuinely sold product, every consumer brand on earth would run an MLM instead of burning billions a quarter on Facebook and Instagram ads for roughly 1% conversion.

Common questions

Are MLMs pyramid schemes?

Regulators have treated them as something adjacent that is harder to prosecute. The FTC shut down Equinox International in 2000 and Trek Alliance in 2005, but the last time it attempted a case at Herbalife's scale was 1979 and it lost. The 2016 Herbalife settlement fined the company $200 million and required at least 80% of US sales to go to genuine retail customers — pricing the practice rather than ending it.

How much do MLM sellers actually make?

Almost nothing, and the distribution is the business model rather than a side effect. At Herbalife 90% of sellers sit in the bottom tier, and the FTC found that those in the middle and bottom earned on average under $5 a month while top sellers grossed over $311,000. At Tupperware the split held for fifteen years — 95% of consultants in the base tier in 2005 and 94% in 2018.

Why did Tupperware go bankrupt?

The sales channel that made it stopped existing. Its advantage was social — a host controlling the venue, hand-picking guests and taking 25–35% of every container sold — which no retailer could copy and no competitor could enter. Once buyers moved online, that structure became a cost rather than a moat. Tupperware went from $4.8 billion in value to bankruptcy in 2024, the same year Avon was dissolved.

What happened to Herbalife?

It saved itself once and could not do it twice. Faced with the same pressures that killed Tupperware, it broke $120 tubs into $2–3 single servings sold from living rooms in Mexico, creating over 20,000 nutrition clubs and rebuilding the company around a new customer. It became the largest MLM on the American stock market by 2015. It still earns billions, but the stock is down more than 80% in a decade and inflation has turned the $9 daily combo into a luxury for its own customers.

Why are MLMs collapsing?

The channel advantage disappeared. Every one of them was built on reaching customers who could not easily be reached another way — suburban living rooms, low-income neighborhoods, communities where information was scarce. Online retail, price comparison and social media removed that gap. Avon was worth $15.5 billion in the early 2010s and is worth nothing; Medifast fell 96% from nearly $300 a share to $11; Nu Skin went from $90 to single digits.

Why don't normal companies use MLM?

Because it does not work as a sales mechanism, which is the most damning fact about it. If a commission structure and a rags-to-riches story genuinely moved product, every consumer brand on earth would use one instead of collectively spending billions a quarter on Facebook and Instagram advertising for roughly 1% conversion. The recruitment, not the retail, is what MLMs are actually good at.

What did the FTC settlement actually change?

Less than the headline suggests. Herbalife paid $200 million and was required to restructure so that at least 80% of US sales go to genuine retail customers rather than to distributors buying inventory. That reshapes the accounting and the compensation plan, but it did not find the company to be a pyramid scheme, and it left the tiered structure and the recruitment pitch intact.

Why did Nu Skin decline?

Its lock-in expired. It sold $300 devices whose gels, cartridges and primers needed replacing every 30 days, so closing one hardware sale created a recurring customer tied to that seller. Through the 2010s and 2020s the beauty market moved to cheaper, more convenient serums and topicals, and the device premium evaporated. The retreat to $20 toothpaste and lip gloss could not replace it — the commission is a fraction of a device system's.

Do MLMs run out of markets?

That is exactly the failure mode. Once a market catches on, the only remaining move is geographic — which is why these companies kept expanding into new countries long after the economics stopped working at home. It buys time rather than solving anything, because the same saturation arrives in each new market on the same schedule.

Discussion

  1. Once the seller and the buyer are the same person, revenue can be manufactured rather than earned. What else in business shares that property, and how would you detect it from the outside?

  2. The model needed people who could not compare prices, could not find the product elsewhere, and had no better option for flexible income. The smartphone removed all three. Which business that you know is currently living on a condition technology is about to remove?

  3. The compensation flywheel cannot be switched off, because the payouts are what hold the sales force together. Describe a commitment an organization you know has made that it can no longer reverse.

  4. If every condition that made MLMs viable is gone, why do they persist? What is actually sustaining them now?

  5. You are a regulator. What single rule would do the most damage to this model — and what would it break that you did not intend to break?

Related case studies